Genotype-Environment Correlation in the Era of DNA
Robert Plomin
0
0
R. Plomin (&) MRC Social,
Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, King's College London
, PO80, De Crespigny Park, London SE5 8AF,
UK
One of John Loehlin's many contributions to the field of behavioral genetics involves gene-environment (GE) correlation. The empirical base for GE correlation was research showing that environmental measures are nearly as heritable as behavioral measures and that genetic factors mediate correlations between environment and behavior. Attempts to identify genes responsible for these phenomena will come up against the 'missing heritability' problem that plagues DNA research on complex traits throughout the life sciences. However, DNA can also be used for quantitative genetic analyses of unrelated individuals (Genome-wide Complex Trait Analysis, GCTA) to investigate genetic influence on environmental measures and their behavioral correlates. A novel feature of GCTA is that it enables genetic analysis of family-level environments (e.g., parental socioeconomic status) and schoollevel environments (e.g., teaching quality) that cannot be investigated using within-family designs such as the twin method. An important implication of GE correlation is its shift from a passive model of the environment imposed on individuals to an active model in which individuals actively create their own experiences in part on the basis of their genetic propensities.
-
John Loehlins influence on my career involves
geneenvironment (GE) correlations of a personal as well as
scientific kind. At the personal level, he introduced me to
behavioral genetics in 1971 when I was a second-year
graduate student in psychology at the University of Texas
at Austin. He contributed to a core course on behavioral
genetics, which included the first Annual Review of
Psychology chapter on behavioral genetics (Lindzey et al.
1971) and was compulsory for all psychology graduate
students. For GE correlation reasons that involve appetite
more than aptitude, this course, and especially John
Loehlins contribution, made me realize that behavioral
genetics was the field for me, even though none of the other
40 students in the core course were similarly enticed to
behavioral genetics.
The beauty and clarity of John Loehlins writing also
attracted me to behavioral genetics. It cannot be a
coincidence that his undergraduate degree was English and that
he is passionate about poetry. In part because of his writing
and the clear thinking that underlies it, his books form part
of the bedrock of behavioral genetics, bringing lucidity to
difficult topics such as race differences (Loehlin et al.
1975), personality (Loehlin 1992; Loehlin and Nichols
1976), and latent variable models (Loehlin 1987). My
favorite is his 1976 book on personality, Heredity,
environment, and personality: A study of 850 sets of twins.
Three quotes from this book illustrate the clarity and lack
of pomposity in his writingas well as the importance of
his findings:
The first clear statement about the importance of
nonshared environment:As far as personality and interests
are concerned, then, it would appear that the relevant
environments of a pair of twins are no more alike than
those of two members of the population paired at
random. Can this possibly be true? (p. 91) Thus, a
consistent though perplexing pattern is emerging
from the data (and it is not purely idiosyncratic to our
study). Environment carries substantial weight in
determining personality it appears to account for at
least half the variance but that environment is one for
which twin pairs are correlated close to zero In short,
in the personality domain we seem to see
environmental effects that operate almost randomly with respect to
the sorts of variables that psychologists (and other
people) have traditionally deemed important in
personality development. What can be going on? (p. 92).
Nearly all psychological traits show moderate genetic
influence (lack of differential heritability): Its message
might roughly be translated: Identical twins correlate
about .20 higher than fraternal twins, give or take some
sampling fluctuation, and it doesnt much matter what
you measure whether the difference is between .75
and .55 on ability measures, between .50 and .30 on a
personality scale, or between .35 and .15 on a
selfconcept composite (p. 35).
One of the earliest multivariate genetic analyses using
twin data: The motivation underlying such analyses is
the hope that they may provide a powerful tool for
studying how genetic and environmental influences
affect phenotypic traits. The basic reasoning runs
something like this: It is unlikely that our convenient
phenotypic trait measures are aligned in a simple
oneto-one fashion with either the genetic or the
environmental sources of influence upon them. If they are not,
the effects of such influences should often show up
more clearly on the associations among traits than on
the measures of the individual trai (...truncated)